Studies of the molecular pathogenesis of prostate cancer (PCA) during the two previous funding periods for this Project have provided new insights into how PCAs arise: somatic inactivation of GSTP1, the gene encoding the pi-class glutathione S-transferase, appears to play a critical role in PCA development. Loss of GSTP1 "caretaker" function increases the vulnerability of prostate cells to genome damage upon exposure to electrophilic carcinogens, like those present in "char-broiled" meats, and to oxidant carcinogens, like those arising as a result of chronic inflammation. Relentless genome damaging stresses over many years, with accumulating somatic genome abnormalities, may be what ultimately leads to life-threatening PCA. For the next funding period, two new hypotheses will be addressed: first, that somatic GSTP1 CpG island hypermethylation changes appear in response to prostate inflammation as a result of targeted repression of GSTP1 transcription and/or increased activity of DNA methyltransferases (DNMTs), and second, that loss of GSTP1 function affects signal transduction pathways in prostate cells to accelerate prostatic carcinogenesis. To approach these hypotheses, four Specific Aims will be pursued: (i) the contribution of transcriptional repression to GSTP1 CpG island hypermethylation will be explored in prostate and liver cancer cells, focusing on the propensity of C/EBPbeta, which appears to trigger GSTP1 trans-repression, to promote de novo GSTP1 CpG island methylation changes, (ii) post-translational mechanisms contributing to an abnormally high DNMT1 levels will be elucidated, (iii) the effects of oxidization and nitrosylation, commonly associated with inflammation, on the enzymatic activities of DNMT1 will be evaluated, and (iv) the contribution of loss of pi-class GST expression to the development of prostate neoplasia in mice will be tested.